Alina retweeted
What is the role of academic computer vision research in the age of increasingly powerful large models? Is GPT-6 Astra a step change? How can a researcher have an impact today in academia?
These are the questions I ask myself as I head off to ECCV 2026, a conference I’ve attended since 1992. One of my papers this year is VIGA, a method that takes an image as input and outputs a 3D Blender scene that represents that image. This is a classical inverse-graphics task and VIGA was the first method to solve it using an agentic approach.
The idea is now several years old and the first version of the paper was rejected. This delayed publication significantly. After it was accepted at ECCV, it was quickly surpassed by people using Claude Code for the same purpose. Today GPT-6 Astra blows away all previous results. But we still head off to ECCV to tell the community about our invention that is now fully out of date.
The way academic work often progresses is that one reads recent papers, notices that they have limitations, comes up with a new idea, explores this, publishes it, etc. Any published paper I read today is based on ideas that are at least a year old. And those ideas were based on the literature of the time, which was also a year old. That means that any paper I see at ECCV is likely two years out of date. In AI today, two years means your work is likely irrelevant.
At CVPR this summer I noticed that many authors have not gotten the message. They continue to work on “old” problems that have a long history. This history is based on assumptions about how the “vision problem” will be “solved”. The truth is that it is being solved in a very different way and many of these problems are no longer relevant. Another group of papers focuses on very niche problems where large models likely fail because of insufficient data or lack of business interest. The impactful papers were largely from industry and had long author lists and massive data+compute behind them. These papers were also out of data, describing systems that had been released months before, but at least they served to provide the community with more complete documentation and analysis of commercial systems.
So what should academics do? First, we need to put aside the tools we’ve used for years and start from scratch. Every project should start by trying really hard to solve the problem with existing tools. I would like to see every paper begin with a detailed experimental analysis of how existing models perform and why they fail (if they do). This gives the kind of insight we need today. Then, assuming current models fail, the solution should provide some fundamental insight that will outlive the next release of such models.
Reviewers today still focus on technical novelty. This pushes people to focus on tweaking architectures rather than clearly moving the field forward. Papers need to be judged based on their novel insight and not their novel technical contribution. This is a real shift in thinking but it focuses us on what matters - progress of the field.
If we want there to be a “field” of computer vision, then it can’t become a marginal backwater, focusing on esoteric problems. If you haven’t tried using Astra (or whatever comes next) to solve your problem, then you have not done your homework. This omission should be seen as negatively as not having a previous work section.
Concretely, I think papers should include a new section analogous to “Related Work” where that related work is current models and how they perform on the task. Reviewers should start asking for this and expecting authors to be able to articulate their insights about the limitations of existing large models.
I'm interested in your thoughts.
Alina retweeted
An engineer that built this 400m-running humanoid robot explained the wild form:
▫️initially designed it with human arm swing
▫️but swinging arms too much put heavy load on shoulder joints, led to overheating (and robot couldn’t finish 400m)
▫️robot invented the new running style on its own (hands held up near face to reduce shoulder load while legs could keep trucking)
▫️per Global Times, “in a simulated learning environment…the team gave the robot continuous rewards, and through repeated iteration on the data it gradually settled into a posture that suited it better.”
***
Full read here (fascinating): globaltimes.cn/page/202608/1…
"Our job is to do as much as necessary and as little as possible" -- Jensen Huang on the Dwarkesh Podscast
27
"Fluency illusion" is a cognitive bias where the brain mistakes familiarity for actual mastery. To break the illusion, you must force your brain to perform the hard work of retrieving information (active recall) rather than passively consuming it.
2
30
Alina retweeted
Attention @arxiv authors: Our Code of Conduct states that by signing your name as an author of a paper, each author takes full responsibility for all its contents, irrespective of how the contents were generated. 1/
Alina retweeted
Today we’re excited to unveil a new generation of Segment Anything Models:
1️⃣ SAM 3 enables detecting, segmenting and tracking of objects across images and videos, now with short text phrases and exemplar prompts.
🔗 Learn more about SAM 3: go.meta.me/591040
2️⃣ SAM 3D brings the model collection into the 3rd dimension to enable precise reconstruction of 3D objects and people from a single 2D image.
🔗 Learn more about SAM 3D: go.meta.me/305985
These models offer innovative capabilities and unique tools for developers and researchers to create, experiment and uplevel media workflows.
Alina retweeted
How Kernel Regression is related to Attention Mechanism - a summary in 10 slides.
0/1
Alina retweeted
"AI isn't replacing radiologists" good article
Expectation: rapid progress in image recognition AI will delete radiology jobs (e.g. as famously predicted by Geoff Hinton now almost a decade ago). Reality: radiology is doing great and is growing.
There are a lot of imo naive predictions out there on the imminent impact of AI on the job market. E.g. a ~year ago, I was asked by someone who should know better if I think there will be any software engineers still today. (Spoiler: I think we're going to make it). This is happening too broadly.
The post goes into detail on why it's not that simple, using the example of radiology:
- the benchmarks are nowhere near broad enough to reflect actual, real scenarios.
- the job is a lot more multifaceted than just image recognition.
- deployment realities: regulatory, insurance and liability, diffusion and institutional inertia.
- Jevons paradox: if radiologists are sped up via AI as a tool, a lot more demand shows up.
I will say that radiology was imo not among the best examples to pick on in 2016 - it's too multi-faceted, too high risk, too regulated. When looking for jobs that will change a lot due to AI on shorter time scales, I'd look in other places - jobs that look like repetition of one rote task, each task being relatively independent, closed (not requiring too much context), short (in time), forgiving (the cost of mistake is low), and of course automatable giving current (and digital) capability. Even then, I'd expect to see AI adopted as a tool at first, where jobs change and refactor (e.g. more monitoring or supervising than manual doing, etc). Maybe coming up, we'll find better and broader set of examples of how this is all playing out across the industry.
About 6 months ago, I was also asked to vote if we will have less or more software engineers in 5 years. Exercise left for the reader.
Full post (the whole The Works in Progress Newsletter is quite good):
worksinprogress.news/p/why-a…
Alina retweeted
One of my favorite formulas is the closed-form of the geometric series.
I am amazed by its ubiquity: whether we are solving basic problems or pushing the boundaries of science, the geometric series often makes an appearance.
Here is how to derive it from first principles:
Alina retweeted
In 15 years as a prof, I've been to 30+ PhD defences.
But only 5 questions determine your success.
The questions seem random, but they're testing skills.
Master these areas and you'll walk out more confident.
Here's the question framework that actually works:
1. Contribution Significance
They want proof your work matters.
This tests your ability to position research impact.
• Start with the broader problem you're solving
• Connect your findings to existing problems
• Quantify the advancement you've made
Move from general significance to specific innovation.
2. Methodological Differentiation
They're probing your research choices.
This shows your understanding of alternatives.
• Acknowledge existing approaches openly
• Highlight your unique methodological angle
• Explain why your approach yields better results
Confidence here shows mastery of your field.
3. Criticism Response
They want to see intellectual resilience.
This tests your ability to defend without defensiveness.
• Acknowledge any valid concerns immediately
• Separate limitations from fundamental flaws
• Redirect to your research's core strengths
Grace under pressure demonstrates research maturity.
4. Future Research Directions
They're evaluating your research vision.
This shows whether you think like a researcher.
• Consider both incremental and breakthrough possibilities
• Connect future work to current limitations
• Identify 2-3 logical next steps
Vision separates PhD students from PhD researchers.
5. Real-World Impact
They want practical relevance demonstrated.
This tests your ability to translate research value.
• Discuss implementation challenges honestly
• Identify specific stakeholders who benefit
• Connect findings to industry applications
Practical thinking proves research influence.
Don't treat these questions as interrogation.
Treat them as your chance to demonstrate expertise.
So:
What question from your defence surprised you most?
Alina retweeted
BREAKING: MIT just completed the first brain scan study of ChatGPT users & the results are terrifying.
Turns out, AI isn't making us more productive. It's making us cognitively bankrupt.
Here's what 4 months of data revealed:
(hint: we've been measuring productivity all wrong)
The UQ AI PhD Showcase is running 26- 27 June 2025. The event features keynotes from Peter Bailey (Canva) & Lawrence Kusz (ChatStat), lighting talks and posters, and a panel discussion on "AI in the real-world" with recent UQ graduates. More info here: ai.uq.edu.au/event/1757/2025…
42
"Anthropic Education Report: How University Students Use Claude" anthropic.com/news/anthropic…
37
Alina retweeted
🧠 What is a Physics-Informed Neural Network (PINN)? 🧠
The problem using purely data-driven approaches can have significant downsides. 🧵👇
Alina retweeted
We’re hiring! @UQSchoolEECS is recruiting a Lecturer/Senior Lecturer (Asst. Prof.) in Applied AI for a continuing teaching & research role. Join us to help build a new Centre of Excellence in AI & Data Science between UQ (Brisbane, Australia) & Thapar Institute (India).


























